Skip to content
Koralis AIBook a free Scoping Review

Kovac · in beta · for finance

AI for your whole workforce, one team at a time.

We want AI on the numbers. We don't want cost prices or customer terms going with them.

Operations wants AI on margin by customer and on the price lists. Finance owns those figures, so finance decides what the AI model may see: the columns you mark in an export reach it masked; board-pack text reaches it as written.

Book a free Scoping Review← Other teams

With both founders, no data. Pilots from £5,000 + VAT + your AI model's usage. Half at the start, and half when we send your payback report.

Day three of month-end, and the board pack is due on Friday. Operations wants the margin-by-customer export in an AI tool to start the commentary. You have said not yet: the same export carries cost prices and every customer's rebate rate.

An illustration, not a customer's story.

One request, from your files to the result

Finance: one request, start to finishYour material: Margin by customer export (a spreadsheet) and Board pack text, as written (a document, by end Oct 2026). Fields you mark: Cost prices, Customer names, Rebate rates. Kovac checks requests to your data against your policy. From late October 2026, the AI model you choose drafts variance commentary for the board pack, working through that check. On every AI call the fields you mark reach it masked: Customer names arrives as [CUSTOMER_1]. Unmarked fields, and the text inside Word and PDF documents, arrive as they are. The result: Draft commentary for finance to review. Separately, today: A manager not cleared for rebates asks for rebate rates, a marked field. Kovac refuses the request and records the refusal. People search that record today; from late October 2026, the AI model can also draft from it. Solid lines work today; dashed lines are dated.Your materialMargin by customer exportBoard pack text, as writtenby end Oct 2026Fields you markCost pricesCustomer namesRebate ratesKovac checks against your policyMarked fields masked on every AI callfrom late Oct 2026The AI model you chooseDraftvariance commentary for the board pack[CUSTOMER_1]in place of customer namesUnmarked fields and Word or PDF text asthey areThe resultDraft commentary for finance toreviewClearance decides what a person may read.A manager not cleared forrebatesasks for rebate ratesRefusedKovac checks against your policy, todayRefused, and recordedA manager not cleared for rebatesasked for rebate ratesPeople search it today.From late Oct 2026, the AI model candraft from it.
  • Works today
  • The AI model's part, from late Oct 2026; Raw files and previews, by end Oct 2026
The detail: the example, what it works on, and what stays governed

Margin by customer, without the customer names or cost prices

An operations analyst asks the AI model why gross margin moved, customer by customer, and for a first draft of the commentary. Finance has released margin by customer and by product, and marked customer names and cost prices. The model receives the released figures as they are and the marked fields as placeholders, finds the movements, and drafts; finance reads the draft before anything is released. A sales manager not cleared for cost prices who asks for that column is refused, and the refusal is recorded.

Fields you mark sensitive reach the AI model masked with placeholders; fields you don't mark reach it as they are. The same value always gets the same placeholder, so masked data can still be linked to a person and remains personal data.

On records loaded into Kovac, someone not cleared for a column you have marked who asks for it is refused, and the refusal is recorded.

Clearance decides what a person may read. It never unmasks a field for the AI model: a marked field reaches the model masked on every call, whoever asks.

What it works on

Document
Board packs and management accounts commentary
Document
Past periods' variance commentary
Spreadsheet
Ledger and trial balance exports
Spreadsheet
Price lists and margin by customer reports, exported as files

Columns this team would mark

  • Cost prices
  • Customer names
  • Customer-specific prices and discounts
  • Rebate rates and payment terms

What the AI does for this team

From late October 2026, the AI model you choose, working through Kovac's check. That model does the work below. Kovac does not draft or search; its check decides what the model may see.

Draft

The AI model does this

Variance commentary for the month-end pack, from the released ledger export, for finance to edit

More it could draft
  • Variance commentary for the month-end pack, from the ledger export and last period's narrative
  • The board pack's narrative, from the figures finance has released

Find

The AI model does this

The product lines that moved gross margin most, from the margin export with customer names masked

More it could find
  • Margin movements by customer or by product, period on period
  • Ledger lines with no matching account in last period's mapping

Check

The AI model does this

A draft board pack's figures against the ledger export, listing any it could not match

More it could check
  • A draft pack's figures against the ledger export, before finance releases it
  • Variance commentary against the movements in the export, listing any explanation it cannot match to a figure

Questions your team will ask

Which of our systems does it connect to?

A pilot works on exports you supply as files: a ledger export, a price list, last period's pack. It does not read from your finance system directly yet. See the limit: when live reads arrive

Who in the business sees the fields we mark?

Whoever your policy clears. Someone not cleared for a marked column who asks is refused, and the refusal is recorded. Clearance governs what people read, not what the model receives.

If we mark margins, can the AI still work on them?

No. A marked field reaches the model masked on every call, whoever asks. Release the margins you are content for it to read, and mark the cost prices and rebates behind them.

Do our figures leave our machines?

Kovac runs on your own machines; where the AI model runs is your choice. A hosted model receives released figures as they are and marked fields masked; your own keeps the request on your network.

What to bring to the Scoping Review

No data: descriptions are enough.

  • The ledger export or price list you would pilot on, and its columns
  • The columns no AI model may see, and who may read each today
  • The first finance task you would give the model, such as variance commentary

What it does not do yet, and what a pilot is

  • Masking works on the fields you mark: the columns of a spreadsheet or record set. It does not reach into the text of a Word or PDF document.
  • The first step is one team: a four-week pilot on your own machines, with one dataset, supplied as files, one workflow, and the AI model you choose working through Kovac's check from late October 2026.
  • Today the check covers three of seven routes to data; opening raw files and previews is checked by the end of October 2026.
The other eight limits

What the check does today

  • Three failure cases still let a request through when they should stop it: a deploy whose policy call fails, a read of a record whose clearance lookup errors, and a plan that cannot be signed, which runs unsigned rather than halting. Every route blocks a request when the policy check cannot run, by the end of October 2026.
  • Kovac's record holds the decisions on the routes it checks: refusals today; permitted requests too by the end of October 2026.
  • The record is searchable, but any signed-in user can add an entry in anyone's name, and entries are not signed or linked together when written. By the end of October 2026, only Kovac itself can write to the record; by then, every entry is signed and linked to the one before it as it is written, permitted requests as well as refusals.

What a pilot is

  • Reading straight from your database tables comes in December 2026.
  • Kovac does not sit in front of Microsoft Copilot, Gemini for Google Workspace or web chat assistants.

What we do not hold yet

  • Kovac is in beta. Koralis AI holds no certification today.
  • Professional indemnity insurance is not held today.
  • A pilot on personal data waits for a data processing agreement, which has no date yet.

Bring the dataset, and the refusal it would have to pass.

Describe the ledger export or price list, and the fields no AI model may see. We tell you what the check would cover, and what not. Both founders, no data, a written pack within 48 hours.

The terms

  • The Scoping Review is free. It signs you to nothing.
  • Four-week paid pilots, with Kovac on your own machines, from late October 2026.
  • Quick-Start, for mid-sized businesses: £5,000 + VAT, fixed.
  • Proof Pilot, for regulated and larger organisations: £7,500 + VAT, fixed.
  • The pilot's job is to give you a payback figure, measured on your own work.
  • Half at the start, and half when we send your payback report. By invoice.
  • You pay for your AI model's usage, either to its provider directly or passed through by us.
  • Stop at the day-five go/no-go and you pay only the first half.
  • If you subscribe within 90 days of the pilot ending, the pilot fee you've paid is credited against your first year's subscription.
  • Your contract is with Koralis AI Limited (company number 16558134).
  • Kovac is not yet in production anywhere.
Book a free Scoping Review

Rather write first? enquiries@koralis.ai reaches both founders. Or message Dylan or Marko on LinkedIn.